Welcome to In the Fast Lane. My name is Edison Yu, and I lead the U.S. Autos, Mobility and Robotics research here at the Bank. We are recording ahead of the Deutsche Bank Autos Conference in New York next week, and we're delighted to welcome Versigent to the program. Joining us from the company, the CEO, Joe Liotine, and the CFO, Doug Ostermann. Thank you both for joining us today. Edison, pleasure. Thanks for having us. Yeah, thanks for having us, Edison. For some background, Versigent is a leading global auto supplier with over 140,000 employees. It generated about $9 billion in revenue last year. Versigent actually has content on one out of every six vehicles made in the world, and within that, one out of every three Battery Electric Vehicles. A very large company with tremendous reach across the globe. To kick off, for those tuning in who aren't as familiar with the company, Joe, could you give us an overview on the main products and offerings you have and generally, who are the competitors we should think about in the same realm? Yeah, sure. Happy to. As you mentioned, we're about a $9 billion business. Principally, our work is to design, engineer and manufacture wire harnesses, that's both low voltage and high voltage. Our revenue really is pretty spread out across the globe evenly, about 40% North America, 25% in EMEA, about a third in APAC. We're principally focused on all vehicles irrespective of powertrain. We're powertrain agnostic. We'll produce and design ICE architectures, hybrid and BEV. Within that, really, we're always looking to design the entire architecture. If it's low voltage or high voltage, if it's distributing signal or data, or power, it really doesn't make a difference to us. We're looking to optimize all of those. From a product portfolio standpoint, low voltage principally on ICE, o n hybrid, you have both, actually on BEVs, you have both as well. Really the only product line part of our business is the charge cords. That's a separate high voltage-only product for BEVs, and for some hybrids. The rest of it really usually is done as overall architecture, kind of mixing both low voltage and high voltage together. Great. I'll just provide a little context. Generally, I think the two competitors that often are brought up are, I believe, Yazaki and Sumitomo. Is that correct? Yeah, the way I would think about it is, the landscape's kind of broken out in two groups, broadly speaking. One group is global competitors, the two you mentioned, as well as Versigent. The remaining, I would say, are more kind of tied to either an OEM specifically or a region specifically, or even a country specifically. They're usually not at global scale, a little bit different characteristics there. The three you mentioned are typically thought of as the global players. Got you. Myself having followed us for a long time, we obviously watched what recently happened with the spin-out. Can you just remind people perhaps not as close to the situation, what was the rationale and what are your top priorities moving forward for the next 12 months? Yeah. Versigent was spun out of Aptiv, greater Aptiv. Aptiv essentially had three main business units, wire harness being one, connectors and some other businesses being the second, and then software and sensing, ADAS, things like that being the third. When Aptiv looked at the overall strategy of the company, looked at the investor base and what they aspired to achieve, frankly, capital deployment, they realized there were pretty different strategies across the businesses, mainly in the wire harness side of things. Essentially we had been, I think, a little bit maybe the second priority, as the company looked to grow in software and other areas. As we discussed how best to really fuel our strategy, it became pretty clear that competing capital priorities was maybe limiting the ability for Versigent to grow. It's been discussed over probably many years on what to do and when to do it. Ultimately, it felt like it was the right time for various reasons to spin off. For us, the exciting part really is a sharpened, more focused strategy regarding wire harnesses, low voltage, high voltage and the future that's coming. A sharper, more deployed resource set, both internally as well as externally. Lastly, just capital. How to put capital behind some of these big ideas that we think are very value-creating, in particular, things like automation and digitalization, where we see really big opportunities. We already have quite a bit of progress in our China manufacturing environment, and we see what we could do with a little bit more precise strategy, a little bit more capital deployed to those ideas. That excites us. I think that really was the reason for the change. We'll continue to look for opportunities within Versigent to deploy capital to the biggest ideas, and we've already kind of demonstrated when doing so, we can generate more growth, and frankly, better margins. That's a pretty big part of our story. Excellent. We'll definitely come back to some of those points you made. Let's start talking about, I think, more about the company and what you're seeing. I think we're in May. A lot has happened this year, I think, in the industry. Can you give us just some insight, I guess, into what you're seeing on the ground, perhaps by region. It seems the dynamics are a bit different. Given your reach, given your scale, and also given the events that have transpired just in the last couple of months, what are you seeing on the ground? Yeah. As you opened with, there's really two or three of us globally that kind of have that same perspective across all the regions. For us, the regional differences probably is the most amount of change we've seen in the last couple of years. What I mean by that is obviously the geopolitical things related to tariffs and potentially conflict and war, but also as it pertains to the knock-on effects to consumers. In the U.S., the EV growth has slowed. It's still growing, but it's slowed versus what people thought in a pretty significant way. Publicly, OEMs made certain disclosures about write-offs and changes of strategy and implications. That was pretty big. I think at the same time, hybrids might be growing a bit faster in North America than people thought as a consequence of some of those things. If you contrast that with Asia-Pacific and specifically China, the BEV movement has not slowed down. I would say it's probably exactly what people thought, maybe even continues to move in that direction. Whereas hybrids might not be quite as strong in its growth as we're seeing. I think that doesn't really surprise anyone, but it is quite different from the two regions. EMEA, I would say, from a powertrain standpoint, probably continuing on the trend they were at, maybe a little different than what people thought, but pretty close, I think. The bigger phenomenon that's taking place in EMEA is the export level from China to EMEA. Really starting probably in Q4 of last year, but it's ramped up and continues to ramp up with really no real end in sight. I would say that's a pretty pronounced change when you're talking about a quarter of the biggest country's production being exported to another region. There's obviously knock-on effects and consequences of that sooner or later will have to come through, both to consumers but also to the industrial footprint. That might then drive another set of either geopolitical things or something else. I think from a market standpoint, that's pretty big. Again, that was supported. We were out in China just a couple of weeks ago at the Beijing Auto Show, and I would tell you it was very consistent from all the local Chinese OEMs. The focus was on both export and on localization in outside markets. That was consistent for everyone. I think when you start talking about the implications of that over a one, two-year period of time, it's pretty significant. From an industry standpoint, those dynamics continue to be quite large. Then I think the tariff question kind of floats in and floats out pretty quickly. So, if that changes again in any market, there's going to be implications of that. Wanted to double-click a little bit on China. You obviously yourself had a very strong quarter there in the first quarter of this year. Can you talk about why you're able to do so well there? On the export side, us being Deutsche Bank, we can see a lot of the Chinese automakers coming to Europe. What do you think the next couple of quarters looks like? Do you think the exports will continue to be quite strong? Yeah. Just from the China market standpoint, we've been in that market decades, so a long time. Our team is really skilled, very much localized to the market, understands the elements and characteristics well, from supply all the way to customer. Separately from China, but more for Versigent, our strategy in market is to match the market. Whatever its composition is, we want to match that, generally speaking, because we know that insulates us from, I'll say, being out of place if things change. For us, we're typically looking in all markets. What is the most complex programs? Where can we add the most value? Where do OEMs need us the most to help in terms of pre-development, development, optimization? We're always gravitating toward the most complex programs. In addition, we want players at scale, and we want players that have global growth potential. Understanding OEM strategies around export and localization is also pretty important. The byproduct of that, the natural selection is there's 100 + labels or brands in China, local Chinese brands. We obviously don't do business with all 110 or 120, whatever it is, because there's high fragmentation. Our strategy helps us get to a natural selection point. We're kind of picking some of these individuals that have the most potential. That played out both last year as well as this year, where, yeah, the market's down almost 20% year-to-date, but we were really working with the biggest OEMs on the biggest programs that actually did the best, both domestically, but also via exports into other markets. We were essentially insulated from a lot of that downward pressure in China from an industry standpoint because of our process, our strategy, who we partner with, and the mix of the programs that we're on. That was a quite helpful thing, and that was both from a customer selection, program selection, and it is a knock-on effect powertrain, right? Since BEVs are the powertrain growing so strongly, BEVs typically have more than 70% more content than an ICE vehicle. There's a little bit of a product mix within the customer and program mix as well. I would just add that our over-indexation with these exporters in China is really a factor of both our strategy and of course, the way in which we are positioned within the market. As we talked about, there's only a handful of truly global players in this market. I think our customers recognize that. Those that are focused on export, they recognize that we can help them as exporters, and then as they localize in those markets around the world, we can also support them there. There are only a couple of competitors who can really provide that type of support, and I think that really makes us quite attractive to those companies. Let's dive a bit more into the operational aspects of the business. I think we've got a pretty good sense on the market. As we know, I don't know if everyone knows out there, wiring, wire harness, notoriously very labor intensive. Can you walk us through the milestones for some of your automation and assembly initiatives? What percentage do you think of your backlog is modeled on these new automated efforts? Yeah. It's a complex topic that has a lot of potential. It's obviously very important to us. The first thing I would say is, automation is a piece of it, but the readiness or the enablers in order to really leverage the benefit of automation happen well beyond that. The engineering design work in terms of bill of process is really important. You can make it as efficient as possible to be manufactured no matter what the design is. That's something we do quite well. In addition, process standards, we have something we call enterprise operating system. How we really just do our work every day, all day long, is really quite robust and it's something that was first established in the Delphi days. We've evolved it since then. Really lastly is automation. If you don't have excellent bill of process and you don't have excellent process work, and you don't have digitalization already contemplated in your plants or in your workflows, then automation is essentially going to automate like a bad idea in isolation as an island. It's really not going to have the network power that it could have if these other things are foundationally there. For us, China's our most advanced location where we do quite a bit of both all the process work and digitalization work, but also the automation work. Really driving toward that is important to make it beyond just theory and beyond a PowerPoint slide, but in practice. Our plant near Shanghai is actually quite automated. We've taken those learnings and started to build essentially like a plan of record of what we want to do globally. We're now rolling into kind of a phasing or a feasibility of how do we roll those things out. It's not as simple as saying we do it in China, so let's do it everywhere else because the business cases, the labor wages, the incentives, the architectures all have different characteristics. You're really optimizing like a set of data to say what makes sense. For us, we're really focused on automation that sits on top of great foundation, A, and B, that has really good paybacks. Two years or less is kind of the general take for us. If we can identify those and execute them two years or less, we know that's going to be value creating for us and value creating for our customers. We're kind of focused there. That roadmap varies a little bit by region. It can vary by product. The complexity of certain architectures have different characteristics on how automatable it is with the right kind of efficiency. All that is contemplated. What we've said publicly is half a point of our two-point margin improvement over the next three years comes from automation. That will vary across the regions. It won't be a static percent in every place. It really is kind of the goal for us. You referenced our, you said backlog, but maybe I'll correct and say bookings because they're actually not orders, but bookings. We can automate low voltage or high voltage. We can automate ICE, hybrid, or BEV. That's not really maybe as big a differentiator as people think it is. They sometimes gravitate toward BEV is the only thing that can be automated, but that's not exactly right. Smaller harnesses can be. The periphery of the plant can be. There's lots of things that can be automated irrespective of the architecture characteristics. For us, we're tackling, I would say, the plant environments first and the architecture second, because, A, we control the plant environment, so we should control our own destiny. Waiting for architectures to evolve or demonstrate characteristics that are automatable, is essentially waiting for someone else to create our strategy. We don't want that. Now, we'll always take advantage of that as architectures evolve, as there's simplicity that gets presented, and if it can be automated, we'll absolutely do that. We don't want to wait for others to dictate our strategy. We're really going to attack the entire perimeter of the manufacturing facility first and foremost. I would say early days today, mostly in China with some specialization in some other areas in Europe and North America, but then a roadmap to get us to this half a point of margin improvement by the end of the three-year window that we laid out. Is there any kind of cool example or interesting example you can maybe help us visualize the automation and work? Anything worth calling out? Yeah. The cool examples depends on what you think is cool. I think all this stuff is cool, so I could do this all day long. I think when you've seen a facility that's high velocity and everything is moving in the facility, like material movement, I mean specifically with AGVs, with big robotic grocery stores, when everything's getting packaged with big transfer lines, it's really quite amazing to see. I think that's more in the general footprint. As it pertains to actual wires, when you see taping automated, it's pretty impressive because the speed with which this thing's happening is quite fast. Even the fidelity, you're talking about very small spaces to engineer robotics in, and they're able to navigate in these really high fidelity spaces and do things very quickly. That's pretty cool to see. I think the other piece I would talk about really is more on the digitalization. What we're able to do now with digitalization and some, I would say, early days AI in terms of sequencing in the plants, really timing all the operations to be as efficient as possible, potentially identifying micro stoppages to really extract out inefficiencies. The amount of power in that data set is not like it ever was before. It would take you days to do some of this analysis. Now it's being done for you. We think the potential there is actually quite significant. Oh, incredible. Okay. Well, yeah. We'll definitely have to get some videos from you or something at some point to showcase. Wanted to move on to the content side. We talked about sort of the automation and some of the initiatives. I think one question that often comes up is, look, there's varying types of content depending on the powertrain, and some are higher and some are lower. How do you think about balancing this going forward? I would also say that in the context of, there are architectures kind of, I think, emerging that are trying to reduce the amount of content but maybe at the value of the content. How do you think about by powertrain content and also some of these trends on next-gen architectures that are looking to, I guess, reduce the weight and the cost of the content? Yeah. It's a really complex, evolving domain area. The thing for us is our differentiation, the reason what makes us great is engineering capability and proprietary engineering toolkits. That really is the reason why we're able to generate almost 2x margins. It's the reason why our revenue, 75% of our revenue is on architectures we influence or help design in some capacity. It's the reason why our growth over market is better than competitors and has been consistently over the last, let's say, quarters and maybe years. That expertise, that capability is about solving problems. Anything that's complex, we don't really care if it's low voltage or high voltage. We don't really care if it's ICE, hybrid, or BEV. Actually, we don't even care if it's on an auto, commercial vehicle, battery energy storage, or robotics. For us, it's about applying a set of capabilities that are uniquely differentiated, that are better than others, that people seek us for, and we're solving their problems and doing it in a way that creates value for our customers and hopefully value for us. That really is the goal. When it comes to content, I think there's some things that are maybe misunderstood. First and foremost, we love complexity, and we want to solve it as smartly and as efficiently as possible. Second, the secular trends are not going to subside. What I mean by that is every vehicle that's launched in the future, doesn't matter what powertrain it is, will have more autonomous features. Period. Every vehicle that's launched in the future will have more in-cabin features in entertainment and connectivity. It's not going to slow down. Every new vehicle that's launched is going to have some migration toward hybrid and EV greater than today. It might vary by region, it might vary by BEV versus hybrid, it's still more. Both those are always more content per vehicle. Second bucket is content per vehicle is only going up. There's no headwinds there. Third is maybe like a 2A, I guess it is. When people talk about optimized, zonal, things like that, a couple of things to understand. One, it takes a lot of engineering to move to those new architectures. Two, they only happen on clean slate, new programs, new platforms. No one's doing that on an existing platform. There's only so many of those that happen each year, every couple of years. They need help to get to those zonal things, which is us oftentimes. We can create a lot of value in that pre-development, development journey with OEMs. What does come out oftentimes is copper, and copper is a pass-through for us. It's not that meaningful to us if we're reducing copper because we don't really get paid on it anyway. Now, there's some small dilutive aspects to it when you do it over time. In general, the tailwinds, combined with our ability to participate in that design, far outpace or exceed any of the potential reductions on content or zonal. Really, I think everyone gets this wrong. The data is super clear. If you look at the data is clear. There's no new architectures that are less, either because of BEV or hybrid or autonomous or in-cabin features. Everything in a car or any vehicle, a sensor, a massaging seat, a screen, any functionality that's put in the car essentially needs either low voltage or high voltage to bring it from A to B. It's not going to get there on its own. Now, you can make it more elegant, you can make it more optimized, but it still needs to go from A to B. We're only creating more and more A to Bs, not less over time. I think for us, and we've depicted this in a couple different examples, but without maybe trying to have a crystal ball to say how much everything grows, because we don't know, but all we know is for sure it's a net tailwind, not a headwind. I think that's really important to understand why our growth over market has been so strong. Well, it's not because we're taking production share. It's because the content's growing at a faster rate than production is growing, and we gravitate toward the most complex vehicles with the most content per vehicle. It's natural to see us grow a bit faster than everyone else. I would only add, Edison, that we do have a number of customers globally who are pretty far down that path of zonal architectures, particularly in China. We have a number of customers who have moved to that, and those are still extremely feature-rich vehicles with a lot of wiring content. Yes. No, I think that was a great breakdown. A couple of follow-ups to that. I think everyone agrees that the feature set of the content is growing. I guess, can you provide maybe some real-life examples of how or instances where, to your point, the wiring or harness kind of goes down, and then you provide an A, B, C, D that actually increases the total value of the program? Is there some easy examples you can maybe bring up? To Doug's point, we're in second, third generation of BEVs that have, I'll say, some zonal aspects. Not complete, not perfect, because there's consequences of that. Those vehicles generate a lot of content per vehicle for us. A lot. They're growing, they're not shrinking. We have other examples in North America, where there's certain aspects of zonal in certain architectures, and yet the content per vehicle is quite high for us, even with those zonals in there. I think, the A to B part is important. What are they optimizing some of these zonals? It might be more processors, it might be more ECUs, it might be more software. You still got to get from A to B. That might be more elegant because at some point you can't just keep stacking one more idea, onto these architectures. It does require a little bit of a re-engineering, a rethink to optimize, but it's still net more. Again, if you look at just the heuristics, hybrids have some optimizations in there. They're 50% more content. BEVs have some zonal and other optimizations in there. They're greater than 70% more content. The role of high voltage is really power. It doesn't do anything besides power. The role of low voltage does everything else, autonomous connectivity, in-cabin, massaging seats, whatever it is. So high voltage as a product is only power. Low voltage is literally everything else. As all these things grow, I think people sometimes take this shortcut of BEV is high voltage. Not really. There's way more low voltage content than there is high voltage content, and they have different roles in the architecture. They're not substitutable. Understood. Okay. Just thinking about, to your point, you have some OEMs in China who have kind of gravitated more to zonal. I think outside of China, maybe outside of Tesla and Rivian, the adoption, I believe, I guess we would call next gen architecture, is still fairly low. Is that picking up? Are you getting a sense that's picking up at all? I think it's going to pick up over time, but again, I go back to, there's only so many clean slate architectures done annually. It's like the opposite of a decay curve, right? There's just only so many that are. There's consequences. It takes a ton of engineering to do that. There are massive knock-on effects to many other things in the vehicle that need to be addressed. Although you can take out weight and copper and some other things, there's some costs as a consequence. More ECU because it's centralized. You need to get everyone on the same software library stack. Just there are other consequences that I think aren't as simple. People get a little enamored with the zonal concepts. As I said, it never happens in isolation. B, there are other very big consequences to those things. Not everyone can do that all at the same time without really big investments. Some of that investment is engineering. Some folks want to spend their engineering on powertrain, or want to spend their engineering on performance. They don't want to necessarily spend all their engineering on what I would say behind the green line architectures, that may or may not be the most compelling feature for that brand to consumers, depending on what the brand is and the consumer use cases. Understood. I want to ask about, you mentioned the copper dynamics, pass-through dynamics, and just more generally, obviously we've seen some pretty big moves in commodity prices. Can you just remind us how the pass-through mechanism work and what are you assuming and what protections you have in place, or hedges you have in place? Maybe I'll turn this one over to Doug, to help with that one. Thanks, Joe. As you mentioned, listen, copper is our largest commodity exposure. Because of that, you'll find that around 75% of the contracts that we have with OEMs have a specific escalation clause within the contract, that increases the price as we see copper step up. The issue that we have there is that roughly, the delay factor, the lag between when the copper price actually moves and when the contract adjustment is made typically is around three to four months. We do see some impact from that. The other 25%, roughly, of our contracts that don't have a specific escalation clause, we do hedge. We hedge over a two-year horizon. Really that just gives us some time to have those discussions with the OEMs about the increase and how it's affecting our cost structure. Those conversations are fairly transparent. They know how much copper is in their product. We know how much is in there. We can observe the various indexes and have a pretty productive discussion around what the impact has been and how to adjust for it, but it takes a little bit of time. That's really what we see impacting kind of the first quarter. We talked about the fact that that was about a $28 million impact in the first quarter on year-over-year basis, some of which we had built into our budget and business plan. We plan for copper to be around $550 on average per pound, for the full year. We'll see how that plays out. We do have hedges in place, and we look at really our hedge strategy, our exposure, our coverage rates, on a regular basis. We've been doing some adjustment as of late. Go through a couple financial questions since we're kind of on that topic, and then have a couple strategic ones after that we can maybe close off with maybe a little fun, maybe a little bit more controversial, I don't know. On the financial one, I think one strong aspect of the story you review, I think your goal is to do roughly $1 billion in cumulative free cash flow by 2028. How do we get there? Are there any certain assumptions that you would highlight that are sort of crucial to hitting that target? We've talked about a couple of things. $1 billion total free cash flow over the three-year horizon. We mentioned that this year will be a bit more muted, kind of in the range of $200 million-$300 million, and that we expect cash flow to, of course, step up over the next two years. A lot of that step-up has to do, frankly, with some one-time costs that we have as a result of the separation. We talked a little bit about the fact that we have about $70 million or so of one-time separation costs that will hit in 2026. Most of that is related to just standing up our own IT systems. That number should drop to about half that figure next year and then disappear altogether. That obviously will impact the progression of cash generation. We've also talked a little bit about the fact that we expect to grow faster than market. We expect that vehicle production, just like IHS, expectations will be a little bit down this year over the three-year horizon be a CAGR of around 1% growth. We expect growth in the 3%-4% range because of all these kind of content per vehicle tailwinds that we just discussed. Of course, we've talked about margin expansion. Joe outlined one of the drivers of that, which is the automation piece, but there are a number of drivers that we think will help expand margins. All that combined makes for a pretty powerful story at the bottom line in terms of top line growth, margin growth, reduction in one-time expenses. You'll see we expect to see a progression in terms of the cash generation over the three-year period. Cumulatively, we're looking at about $1 billion. We have some debt holders who probably want me to ask this leverage. I think you launched or you came out, I think you had to do about $2 billion, a little over $2 billion to fund the dividend to Aptiv. How does one think about the leverage ratio and how aggressive you'll be with kind of early debt retirement versus reinvestment? Yeah, right now we feel pretty comfortable with the capital structure that we have in place. I think historically, there have been spins that have been heavily laden with debt. That's not the case with Versigent. We have a pretty healthy balance sheet, I would say, a nice credit rating as a highly rated, high yield name. We're pretty comfortable with the ratio when you look at net debt to EBITDA at about two times. We've talked about our gross debt to EBITDA ratio that we'd like to maintain between two and two and a half times. I think that metric will improve over time. One, because we'll be growing EBITDA, and two, because we do have a little bit of natural pay down in the debt structure. Part of the structure is a $500 million TLA that has an amortization of about $100 million over a number of years. We'll have a little bit of debt pay down that is natural. I think we feel pretty comfortable with the strength of the balance sheet at this point and where we sit from a credit ratings perspective. I don't think there's a lot of work to be done there. I guess, the message to the investment community is, at this point, I don't see a lot of change in terms of the ratios that we're talking about in terms of leverage and that sort of thing. Great. Want to conclude with a couple of strategic questions. First, both of you have been in the industry a long time. I remember maybe 10 years ago, there was talk about, oh, we need more consolidation in electrical architecture. Maybe some of the regional players. I guess what's the latest thinking on just the industry dynamics? Have we consolidated enough now? Do you think there's more coming? If so, would you want to be the one to initiate that? Curious your thoughts there. I don't have a crystal ball on how others view it or what will transpire. I think there has been some consolidation. Not so much maybe in Americas and EMEA, but more on the APAC side of things over the last couple of years. For us, we look at ourselves as we're one of the best operators. We're growing faster than others. We're focused on our strategy. We're focused on fueling that. I think there's implications if we're as successful as we'd like to be, if our competitive advantage really resonates the way we think it should and does. For us, I think consolidation is probably more a priority for others than it would be for us to talk about because we're winning, and we think we have even more value we can create once we start fueling these strategies that maybe were a little bit under-fueled in the past. I wanted to ask about some of the secular markets outside of autos. I think you mentioned battery storage before. You also had some announcements, I think, about robotics. How does one think about those opportunities and the timeline into making inroads there? Yeah. As we shared as we opened, about 10% of our revenue comes from non-auto, and much of that is commercial vehicles, heavy equipment on-road, off-road, agriculture. Growingly, we've better understood our engineering capabilities fully apply in those other sectors. They don't require any real change or incremental investment. It's really just a requirements or input change for us, which is easy to manage. Our manufacturing environments also apply quite well without big changes to that either. Really for us, it's really about go to market and sales, which is a smaller, quicker kind of investment for us to really build up the team to better know the players, to better know the process, and really be a lot more proactive. Historically, of that 10% revenue, all of it was essentially reactionary. Someone came to us and asked us to do it. What if we were actually proactive? What if we were actually focused? What if we put resources behind it? What could it be? I'm assuming bigger than 10%, if 10% is just us fulfilling requests. We're focused there. It's early days. Our first serial battery energy production was in Q1. We're in pre-development on some more battery energy storage, and we're in pre-development with a few robotics players as well. We think those areas will grow. The commercial vehicles probably an easier extension for us because we're already doing it today. These other areas are smaller, more nascent sectors, but their CAGRs are quite strong. We're excited about that complement to our overall revenue profile for the future. Fantastic. I think we could talk on for another hour, but I appreciate both of you joining us today. It's been a great session. Until next time, thank you again. Thank you. Good to see you again. Thank you. Thanks, Edison. Yep. Take care.